Rationale: For resectable esophageal cancer the standard therapy is 5 weeks of neoadjuvant chemoradiotherapy (nCRT) followed by surgery 6-8 weeks afterwards. Surgery is performed independent of the response to nCRT and is associated with substantial morbidity. A pathological complete response (pCR) after nCRT is seen in 28-34% of patients. Pathological non-responders (pNR) most probably do not benefit from nCRT but are exposed to its toxicity and delay from surgical therapy inevitably occurs in this group. Early identification of non-responders during nCRT would allow individualized decision making in continuation or discontinuation of nCRT. Furthermore, a tool is desirable to accurately assess the treatment response after nCRT to identify patients with a complete response. Studies in rectal cancer reported that tumor resection could be omitted in patients with persisting clinical complete response after 12 months. Also, in some esophageal cancer studies, complete responders in surgical and non-surgical treatment groups had comparable overall survival. These findings indicate the possibility to perform nCRT as sole treatment in patients with a complete response. Conversely, if residual tumor is demonstrated, this will support the decision to move to surgery.

Objective: Diagnostic study to assess the distinct and combined value of anatomical and functional magnetic resonance imaging (MRI) and combined 18F-fluorodeoxyglucose positron emission tomography and computed tomography (PET-CT) in the evaluation of treatment response to nCRT for patients with esophageal cancer.

Study design: Multi-center diagnostic study investigating the value of MRI and PET-CT in the imaging before, during and after nCRT for assessment of response to nCRT for resectable esophageal cancer. Imaging response measurements will be compared with the histopathological tumor regression grade (TRG) of the resection specimen as gold standard.

Study population: 50 patients (>18 years) presenting at the UMC Utrecht or M.D. Anderson Cancer Center with resectable esophageal cancer, as determined by endoscopy and biopsy, computed tomography (CT) and endoscopic ultrasonography (EUS), receiving nCRT prior to surgery.

Procedure: In addition to the conventional diagnostic work-up for esophageal cancer including a standard PET-CT before nCRT, two PET-CT scans will be performed during and after nCRT as well as three MRI scans before, during and after nCRT at the same time points. The first MRI (and standard-of-care PET-CT) scan session will be within the 6 weeks prior to the start of nCRT. The second scan session will take place after 10-14 days after the start of nCRT. The third and final scan session will be planned 1-2 weeks before surgery.

Main study parameters/endpoints: Determination of the distinct and combined diagnostic value of anatomical and functional MRI and PET-CT in the evaluation of treatment response to nCRT for patients with esophageal cancer. Important imaging parameters include apparent diffusion coefficient (ADC) values, standardized uptake values (SUV) and volume measurements for the different time points. The differences of these values between time points are of particular interest (delta-ADC, delta-SUV, delta-volume). The sensitivity (%), specificity (%), negative predictive value (%), positive predictive value (%), and accuracy (%) of the different imaging parameters for correctly identifying histopathological complete response will be calculated.

Multi-center diagnostic study investigating the distinct and combined value of MRI and PET-CT for response prediction before and response assessment during and after nCRT for esophageal cancer.

The patients will be enrolled in two hospitals (UMC Utrecht and M.D. Anderson Cancer Center) during their first outpatient visit at the Department of Surgery or Radiation Oncology. At the UMC Utrecht, for study purposes three anatomical and functional MRI scans are made and two PET-CT scans, in addition to a pre-nCRT PET-CT scan which is standard care. At M.D. Anderson Cancer Center, for study purposes three anatomical and functional MRI scans are made and one PET-CT scans, in addition to a pre-nCRT and a post-nCRT PET-CT scan which are standard care. The MRI scan protocol will consist of anatomical and functional MRI scans. For the dynamic contrast-enhanced MRI (DCE-MRI) scans, a contrast agent will be administered to the patient. All MRI scans will be performed on a 1.5-Tesla MRI scanner. For the PET-CT scans, 18F-fluorodeoxyglucose (FDG) is the tracer that will be used for the assessment of abnormal glucose metabolism in the tumor.

Before the start of nCRT, we schedule an MRIpre to have baseline MRI and PET-CT images in all enrolled patients. This scanning day is matched to the irradiation treatment planning PET-CT scan, which is routinely made for treatment planning purposes. Next, the patients receive standard nCRT treatment with weekly administration of carboplatin (doses titrated to achieve an area under the curve of 2 mg per milliliter per minute) and paclitaxel (50 mg per square meter of body-surface area) for 5 weeks at UMC Utrecht and 6 weeks at M.D. Anderson Cancer Center with concurrent radiotherapy (41.4 Gy in 23 fractions in Utrecht or 50.4 Gy in 28 fractions at M.D. Anderson, delivered 5 days per week on workdays). As mentioned in the introduction, earlier studies reveal that both PET and DWI-MRI after 2 weeks of treatment showed the best distinction between responders and non-responders. Therefore, we decided to perform a PET-CTper and an MRIper scan two weeks after the start of the nCRT. Surgery will follow 6-8 weeks after finishing nCRT and all patients will undergo a preoperative MRIpost and PET-CTpost scan 1-2 weeks before surgery, combined with a standard preoperative outpatient follow-up appointment. Overall, no extra hospital visits are asked from participants as we try to schedule all study activities at days corresponding with appointments for standard care. If in individual cases an extra visit to the hospital is needed, travel expenses of this extra visit will be reimbursed.

Important imaging parameters include apparent diffusion coefficient (ADC) values, standardized uptake values (SUV) and volume measurements for the different time points. The differences of these values between time points are of particular interest (delta-ADC, delta-SUV, delta-volume). The sensitivity (%), specificity (%), negative predictive value (%), positive predictive value (%), and accuracy (%) of the different imaging parameters for correctly identifying histopathological complete response will be calculated. The resection specimen will be evaluated meticulously according to standard protocol (tumor type and extension, lymph nodes, resection margins), and additionally, findings of effects of nCRT will be reported.

The research population consists of patients with resectable esophageal carcinoma, scheduled to receive preoperative chemoradiation according to the CROSS-regimen. The participants will be recruited from the surgical outpatient clinic or radiation oncology department of the UMC Utrecht or M.D. Anderson Cancer Center. In total 50 participants with available histopathologic data of the resection specimen are required for this study.

Inclusion criteria: In order to be eligible to participate in this study, a subject must meet all of the following criteria (comparable with CROSS-criteria):

• Histologically confirmed squamous-cell carcinoma, adenocarcinoma or large-cell undifferentiated carcinoma of the esophagus or esophagogastric junction (i.e. tumors involving both cardia and esophagus on endoscopy)
• Potentially resectable tumor (cT1b-4a N0-3 M0): based on standard primary staging by EUS and CT
• Undergoing preoperative chemoradiation according to CROSS-regimen
• Age>18 years
• No history of other cancer or previous radiotherapy or chemotherapy
• Signed informed consent

Exclusion criteria: A potential subject who meets any of the following criteria will be excluded from participation in this study:

• Patients who meet exclusion criteria for MRI at 1.5T following the protocol of the department of Radiology of the UMC Utrecht
• Patients who meet local exclusion criteria for intravenous contrast (Glomerular Filtration Rate (GFR) of <45 mL/min/1.73m2, unless the patient has risk factors for contrast nephropathy according to the UMC Utrecht protocol 'Prevention contrast reaction and contrast nephropathy, Version 2 February 2012'). In patients with risk factors (2 or more of the following: 'peripheral vascular disease, heart failure, age >75 years, anemia, symptomatic hypotension, dehydration, use of diuretics or non-steroidal anti-inflammatory drugs)', a minimum GFR of 60 mL/min/1.73m2 will be required
• Patients with insulin dependent diabetes mellitus or blood plasma glucose concentration higher than 10 mmol/L
• Patients with a known Gadovist allergy
• Patients with a known CT-contrast allergy
• Patients having difficulty understanding Dutch
• Pregnant or breast-feeding patients

Inclusion Criteria:

• Histologically confirmed squamous-cell carcinoma, adenocarcinoma or large-cell undifferentiated carcinoma of the esophagus or esophagogastric junction (i.e. tumors involving both cardia and esophagus on endoscopy)
• Potentially resectable tumor (cT1b-4a N0-3 M0): based on standard primary staging by EUS and CT
• Undergoing preoperative chemoradiation according to CROSS-regimen
• Age>18 years
• No history of other cancer or previous radiotherapy or chemotherapy
• Signed informed consent

Exclusion Criteria:

• Patients who meet exclusion criteria for MRI at 1.5T following the protocol of the department of Radiology of the UMC Utrecht
• Patients who meet exclusion criteria for intravenous contrast (Glomerular Filtration Rate (GFR) of <45 mL/min/1.73m2, unless the patient has risk factors for contrast nephropathy according to the UMC Utrecht protocol 'Prevention contrast reaction and contrast nephropathy, Version 2 February 2012'). In patients with risk factors (2 or more of the following: 'peripheral vascular disease, heart failure, age >75 years, anemia, symptomatic hypotension, dehydration, use of diuretics or non-steroidal anti-inflammatory drugs)', a minimum GFR of 60 mL/min/1.73m2 will be required
• Patients with insulin dependent diabetes mellitus or blood plasma glucose concentration higher than 10 mmol/L
• Patients with a known Gadovist allergy
• Patients with a known CT-contrast allergy
• Patients having difficulty understanding Dutch
• Pregnant or breast-feeding patients

The research population consists of patients with resectable esophageal carcinoma, scheduled to receive preoperative chemoradiation according to the CROSS-regimen. The participants will be recruited from the surgical outpatient clinic or the radiation oncology department of the UMC Utrecht or M.D. Anderson Cancer Center. In total 50 participants with available histopathologic data of the resection specimen are required for this study.